USDA;'FAER-2S2 ENUIRON"ENTAL. .POL IC IES: I "PL I CMi:ONS:FO~.~~;iJ AGRICULTURAL TRADE . (AGRICULTURE ECONO"IC REPT.(FINAL>.) . ;' J.SULUUAN ECONO"IC RESEARCH.. SERUICE. WASHINGTON. DC. AGRICULTUREAN~ TRA~EANALYSIS DIU. JUN 94 lS3P

,.'

~ rt

\ L AlIM if ~ i~~

....c..._to"nfonn.._.nd'~~~ i~~ 1100 W mage Management ~ // i~~ ~-2 Silve ayn,eAvenue,Suile1100 ~~.I ~~;!f .I$"~rSpnng'Maryland20g10~' ~...." .qr",U".....

3011587-8202 . -I!

III III PB94-188703

Environmental Policies: Implications for Agricultural Trade

Economic Research Service, Washington, DC

Jun 94

,ll

[ 'm

I

u.s. DEPARTMENT OF COMMERCE Natioml.l Technical Information Service

N'nS

Form APP1'OVOOREPORT DOCUMENTATION PAGE OMB No. 0704''()188 Public ..poctlnglOr lhI. coiiec.loc>oFlafnrmllloc> 1....lm>led 10 .,..." I !.our por ...poo lridudlng.he .Im. lOr ..""wine "uorue.1oru, .....-chtngubtln'd>O. """"" .. pili.rtne and maintain'", u.. ch needed. and eemple.lng and ..""wine u.. ccIIec,loc> oFlnI'''''''atlon. S.nd eemm .,: ...go.rtiln, ,hi. butdoo ."Im '" ""Yother "P"" oF,hI. ccIIec.loc>oFlnI'''''''>ltoo.1ndudlnC lUU."Ioc>.lOr ..duel", lhI. butdon.1O Wuhln~on Ho.dquulI.. SeMce" DIM'","', lOr InI'nrm>llon Opo...Ioc>nd p.,pocu. IllS ]t1litrl"" D .,;. Highwoy. Sul 1l0'.ArlIngtoo. VA =2.-'.302..nd.o u.. Oll1c. oFM..,.",mtn. and Bud",. P>porwor~ p.,due.1on Pcoj.e. (07M-OI88). WuhlnOlon.DC 2Il~n~

2. REPORT DATE 3. REPORTTYPEAND DATES COVEREDIII 1111 II lill 11111111111 1111 III

PB94-188703 June 199Jf,.. final 4. TITLE AND SUBTiTLE

'

Environmental Policies: Implications for Agricultural Trade

6. AUTHOR(S)

John Sullivan

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES), Economics Research Service U.S. Department of Agriculture

Washington, DC 20006-4788

SPONSORING/MONITORING NAME(S) AND ADDRESS(ES)

II. SUPPLEMENTARY NOTES .

12a. DISTRIBUTION!AVAILABILITI STATEMENT

.>

13 .~ABSTRACT (Maximum 2(10 words)

5. FUNDING NUMBERS

8. PERFORMING ORGANIZATION REPORT NUMBER

10. SPONSORING/MONITORING AGENCY REPORT NUMBER

FAER-.;?52

\

112b. DISTRIBUTION CODE

I THis""report, consisting of 14 separate articles, analyzes linkages between environmental policies and agrIcultural trade. Topics covered include: a global inventory of environmental policies, the implications of envir~f;jinental policies on U.S. and world agrcultural trade, the implications for environmental policy in the context of multilateral and regional trade negotiations, and the efffects of global climate change on agricutural trade.-{ ..

14. SUBJECT TERMS 15. NUMBEROFPAGES

(to be decided by author) 236 16. PRICE CODE

:7. SECURITICLASSIFICATION 18. SECURITY CLASSIFICATION 19. SECURITI CLASSIFICATION 20. LIMITATION OF OF REPORT OFTHISPAGE OF ABSTRACT ABSTRACT SAR

Unclassified Unclassified Unclassified ~. ~

Standard Form 298 (Rev. 2-89) P,.scribed by ANSI SId. ,.39-18

198-10l

http:WuhlnOlon.DChttp:2.-'.302..ndhttp:butdon.1Ohttp:lridudlng.he

~; :~(lr.'l'1'T'r;n! i~f 1)1"1""''1''0'''0

~~"t; )",1 I"n~"

7 r F

It's Easy To Order Another Copy!

Just dial 1-800-999-6179. Toll free in the United States and Canada. Other areas, please call1-703-834-012S.

Ask for Environmental Policies: Implications for Agricultural Trade, FAER252. (AER-689).

The cost is $15.00 per copy ($18.75 for non-U.S. addresses, including Canada). Charge your purchase to your Visa or MasterCard. Or send a check (made payable to ERS-NASS) to:

ERS-NASS 341 Victory Drive

Ht3rndon, VA 22070

.M

The United States Department of Agriculture (USDA) prohibits discrimination in its programs on the basis of race, color, national origin, sex, religion, age, disability, political beliefs, and marital Dr familial status. (Not all prohibited bases apply to all programs.) Persons with disabilities who require alternative means for communication of program information (braille, large print, audiotape, etc.) should contact the USDA Office of Communications at (202) 7205881 (voice) or (202) 720-7808 (TOD).

To file a complaint, write the Secretary of Agriculture, U.S. Department of Agriculture, Washington, DC 20250, or call (202) 720-7327 (voice) or (202) 720-1127 (TDD). USDA is an equal employment opportunity employer.

Environmental Policies: Implications for Agricultural Trade. Edited by John Sullivan. Agriculture and Trade Analysis Division, Economic Research Service, U.S. Department of Agriculture. Foreign Agricultural Economic Report No. 252.

Abstract

This report, consisting of 14 separate articles, analyzes some of the linkages between environmental policies and agricultural trade. Topics covered include: a global inventory of environmental policies, the implications of environmental policies on U.S. and world agricultural trade, the implications for environmental policy in the context of multilateral and regional trad.:: negotiations, and the effects of global climate change on agricultural trade.

Keywords: environmental policies, agricultural trade, inventory and analysis, trade negotiations, global climate change

Acknowledgments

Many people in both the Agriculture and Trade Analysis Division and Resource and Technology Division contributed to this report. Particular thanks go to the following reviewers: Nicole Ballenger, John Dunmore, Ken Forsythe, David Skully, Martin Johnson, Barry Krissoff, Rip Landes, Carl Mabbs-Zeno, Mary Lisa Madell, Tim Osborn, Steve Magiera, Steve Haley, Dale Leuck, Donna Roberts, and Denice Gray. The review of Phil Shull of the Foreign Agricultural Service was also appreciated. The editing of Carol Morgan is acknowledged, as well as the clerical support of Verleece Hill. Camera copy was prepared with the expert desktop publishing advice of Linda Ghelfi.

Washington, DC 20005-4788 June 1994

iii

Contents

Page Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1 Introduction ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. by John Sullivan

1

2 Environmental Accounting: Including the Environment in Measures of Well-Being ......................................... ,. by Rachel Beattie

6

3 Measuring the Impact of Government Policies on the Environment by Carl Mabbs-Zeno and John .Antle

....... 13

4 Policy Choices in the Use of Land and Water by Denice Gray and Gary Vocke

...................... 20

5 Multilateral Environmental Activities by Carl Mabbs-Zeno

.......................... 29

6 Environmental Standards and Regulations in a Global Context by David Skully

. . . . . . . . .. 39

7 Agricultural Trade and the Environment: Some Basic Economic,Principles .......'....................... 48 by Denice Gray and Barry Krissoff

8 NAFT A and Environmental Quality: by Margot Anderson

Issues for Mexican Agriculture . . . .. 58

9 Coordination of Agricultural Trade and Environmental Policy: A Western Hemisphere Example ................. ,............ by Nicole Ballenger, Rachel Beattie, and Barry Krissoff

75

10 Costs and Benefits of Irradiation Quarantine Treatments for U.S. Fruit and Vegetable Imports. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. by Kenneth Forsythe and Phylo Evangelou

82

11 The EC Nitrate Directive and its Potential Effects on EC Livestock Production and Exports of Livestock Products . . . . . . . . . . . . . . . . . . . .. by Dale Leuck

91

12 Assessing Environmental and Agricultural Policy Linkages in the European Community ................................ by Stephen Haley

102

13 Tropical Forest Policy and Trade: by Mark Giordano

A Case Study of Malaysia . . . . . . . .. 113

14 Global Climate Change: Analyzing Environmental Issues and Agricultural Trade within a Glbbal Context ............ , . . . . .. by Roy Darwin, Jan Lewandrowski, Brad McDonald, and Marinos Tsigas

122

ii

Summary

This report explores data needs and indicators, examines the institutions addressing trade and environment issues, and begins an analysis of the implications for agricultural trade of changes in environmental and agricultural policies. Examples of modeling exercises are presented, in an attempt to bring into focus some of the many possible interactions between trade and environmental policies. This report also includes surveys and case studies of Issues related to trade and the environment.

Following the introductory chapter, thld first five papers (Chapters 2 through 6) 'focus on environmental policies and standards and on institutions governing and affecting global environmental issues. Beattie compares economic measures of living standards across countries, including environmental degradation. More specifically, she eAamines efforts to include price measurements of "environmental depreciation" in the System of Nationa! Accounts, in response to charges that traditional measures of economic growth undervalue the natural resource base and overestimate social welfare. Mabbs-Zeno and Antle layout a theoretical basis for quantifying the environmental impacts of policies designed principally for agricultural purposes. The article by Gray and Vocke, based on the forthcoming Global Review of Resource and Environmental Policies, provides a status report on environmental policies relating to land, water, agricultLlral chemicals, and wildlife for about 25 countries. Gray and Vocke concentrate on water and land policies. '

Mabbs-Zeno describes the programs of multilateral institutions directed toward environmental issues and how international lending policies have changed to reflect environmental concerns. Skully continues the discussion on multilateral institutions, emphasizing the role that international organizations may have in resolving disputes over product and process standards. The internationc.l institutions that attempt to mediate, regulate, and standardize environmental and health standards are surveyed as to how they resolve these issues.

The rest of the report consists of papers on environmental policies and their potential effects on trade. These start with the paper by Gray and Krissoff, which lays out basic economic principles for use in analysis of agricultural trade and the environment. The following articles take a regional perspective: three papers examine trade and environment issues relating to Western Hemisphere concerns, two papers focus on the European Community's environmental policies and their effect in EC and world agricultural trade, and one article addresses Malaysia's ability to sustain its exports of tropical lumber.

Anderson's article discusses the impact of NAFTA-induced freer trade on longerterm agricultural environmental issues, particularly for Mexico. The article discusses agricultural environmental concerns and issues in Mexico and the United States, and projected environmental consequences of NAFTA. Ballenger, Beattie, and Krissoff look at potential agricultural trade agreements in the Western Hemisphere. This issue is explored in a conceptual model and in empirical case studies. In the conceptual model, the authors find that regional trade agreements may set the stage for a bargain between groups in the "North" representing environmental interests, and groups in the "South" representing trade expansion interests. The case study considers a preferential trade agreement for a single commodity and the regulation of an environmental externality involving pesticides associated with the production of that good. In the third article related to Western Hemisphere agricultural trade and the environment, Forsythe and Evangelou examine a possible future ban on the use of methyl bromide for treatment of imports.

iii

Leuck analyzes the effects of the passage of a regulation by the EC environmental ministers that limits the application of nitrogen-containing manure and chemical fertilizers in areas where nitrate levels are already high. Some areas of the EC are to be designated "vulnerable zones" where fertilizer and/or manure application must be reduced. Restrictions on livestock density in these regions are expected. Haley's article includes a description of EC environmental problems and their relationship to agriculture, with an emphasis on the role of EC policies in encouraging the intensification of both livestock and crop agriculture. Several new directions in EC agricultural and environmental policies are considered.

Giordano's article focuses on Malaysia, but takes the form of a case study of a sustainable agriculture issue. Malaysia possesses some of the largest tracts of tropical forests in the world, and is the world's largest exporter of both tropicc.ll logs and sawn timber. However, growing realization of the environmental degradation caused by deforestation and a recognition that logging rates of past decades have been above those sustainable on a long-term basis have prompted changes in both timber extraction and export policies. Recent policy innovations have caused a decline,in timber extraction rates and a radical shift in the composition of exports toward relatively high-valued wood products.

The report concludes with an assessment of the potential effects of climate change on agricultural trade. While the magnitude of the global problem is debatable, many agree that significant regional agricultural problems are possible. The final article by Darwin, Lewandrowski, McDonald, and Tsigas reinforces this point. LookinfJ at shifting uses for natural resources in a changing climate, these authors find that improvements in analytical capacity to assess potentiai geographic shifts in land classes allow them to pinpoint some potentially disturbing shifts in agricultural production.

iv

Environmental Policies Implications for Agricultural Trade

Chapter 1

Introduction

John Sullivan 1

The global community has recently become more concerned about the environment, making it one of the leading issues of the 1990's. Many of these issues, which include water quality, soil erosion, deforestation, pl'Oduct safety, and protection of wildlife and biodiversity, and the policy measures adopted to deal with them, relate closely to agriculture and agricultural trade. For analysts of agricultural trade, the environment may be the most important trade issue of the 1990'S.2

This perception of environmental importance follows several long-term deveiopments strengthening the threat from environmental degradation. As the global community responds to these problems, it challenges existing international arrangements to absorb an additional mandate, sometimes forcing compromise on such traditional priorities as commercial trade in favor of sustainable production and safe consumption.

International environmental protection began concurrently with domestic programs nearly 150 years ago as colonial administrations recognized the dramatic impact they were having in less developed areas of the world. The British passed legislation to protect open areas near Cape Town in southern Africa in 1846. Also, acts to protect forests and game in southern Africa were passed in 1859 and 1886. In comparison, the first national park in the United States (Yellowstone) was formed in 1872. The seven Europoan countries with African colonies met in London in 1900 to sign a convention for animal protection in Africa. A broader environmental agenda was pursued by industrialized nations in 1909 at the International Congress for the Preservation of

'The author acknowledges Steve Haley, Barry Krissoff, and Carl Mabbs-Zeno for their extensive contributions to this chapter.

2For instance, see the comments, cited in the References, made by Michael Smith, former Senior Deputy U.S. Trade Representative and U.S. Ambassador to the GATT.

Nature. By the mid-1930's several international governmental organizations had been formed for environmental protection.

A new approach to international environmental action emerged during the 1960's from developments in the science of ecology and from the rapid extension of transportation networks that linked countries more closely. These efforts did not result in strong international commitments, however. The Stockholm Conference of 1972 represented broad recognition of environmental issues, but it resulted in little concrete change of policy.

A variety of international agreements have followed to address the use of natural resources that are shared internationally. Numerous river basin agreements and the Montreal Protocol, which established rules for use of the atmosphere, took effect after the Stockholm Conference. The recent strength and breadth of international involvement in environmental protection was established financially by the Global Environment Facility (GEF) in 1990 and politically by the U.N. Conference on Environment and Trade held in Rio de Janeiro, Brazil, in 1992 (the Rio Conference). The GEF has already disbursed $1.3 billion through the World Bank and the U.N. Environment Programme. The Rio Confert.')nce attracted over 150 nations to the largest meeting ever held to promote international agreement on. environmental protection.

Environmental problems attract international concern when production activity in one country affects the emiit"onment in another country, or when controls on production activity affect competitiveness among countrielS. Environmental damage across national borders occurs when pollution damages a neighboring country or the global commons. Localized agricultural examples include runoff of chemicals onto neighboring

1

countries and the diversion of :rrigation water that otherwise would have crossed a border. Global examples include the effect on the atmosphere of land-clearing operations that increase carbon dioxide production and decrease oxygen production. Nitrous oxide and methane are byproducts of agricultural production that could contribute to a longrun warming of the world climate.

Domestic concerns for environmental quality can also impact on global markets for agriculture, fisheries, and forestry. Two very different examples have recently received public attention: the directive of the European Community (EC) to reduce surface and groundwater contaminants through spillover and leaching; and the U.S. restriction on yellowfin tuna imports caught by purse seine fishing nets. Each of these actions represents a government policy to eradicate a perceived market failure--a failure to adequately value an environmental good, namely water quality and animal life (dolphins). Each represents an internationalization of environmental policies. In the EC case, the environmental policy, in effect, limits fertilizer use and implies a change in production techniques, food output, and ultimately, changes in trading patterns. In the U.S. case, the objective of saving the lives of dolphins leads to a trade instrument, also altering trading patterns. Other nations in Europe and Asia also have joined the U.S. boycott.

The shift in international attention to better account for environmental issues inevitably confronts traditional motivations underlying trade policy formation. Trade itself is an instrument for (or means of) enhancing national welfare through maximizing a nation's income and growth potential. Especially since the end of the Second World War, trade has played a large role in motivating international cooperation. Institutions designed to promote or control trade are now adjusting by incorporating explicit environmental considerations. In 1991, the Trade and Environment Group of the General Agreement on Tariffs and Trade (GATT) met for the first time since it was formed in 1971. Trade institutions are motivated in this adjustment both by a recognition of environmental effects of policies designed for trade purposes and by trade effects of policies designed for environmental protection.

This report examines a critical intersection of concerns by focusing on issues that are simultaneously environmel')tal, trade-related,

agricultural, and international. It provides articles ranging from the introductory theory underlying policy design to case studies of specific policy reforms. This range is too broad to attempt comprehensive coverage of the topics, but the report is made practical bV its convergence on issues of immediate interest: approaches to free trade, European policy reforms, and global warming.

Clarifying the Trade-Environment Nexus

Trade occurs because of its capacity to improve national welfare through consumption of a wider variety of products. The concept or practice of trade and maintaining environmental quality are comparable in the sense that they share the same goal: improved social welfare. The nexus or conflict occurs where economic activity is conducted in such a way or scale as to harm the environment, and where environmental regulations inhibit economic activity, such as trade. Environmental quality, although more intangible, can be considered in several ways, including: something that can be consumed (such as recreation), a more healthful work setting, or the setting in nature in which consumption takes place. In this last sense, a pleasant or nonharmful setting contributes to the utility derived from the consumption itself.3

Partially due to these conceptual difficulties, the debate over how to adjust policy for trade and environmental problems has been hampered by ambiguity in the use of these terms. Both terms are used to define policy purpose, instruments, and effects. Figure 1 provides a framework for delineating the issues in the trade-environment nexus.

The first column in figure 1 represents trade and environmental goals, and possible tradeoffs between them. Serving one type of goal often serves the other as well. For example, improved environmental quality might improve agricultural production and trade. In other circumstances, however, there is a tradeoff between trade and environmental goals. Questions then arise of how much loss in environmental quality is acceptable

3A conceptual difficulty that has masked the contribution of environmental quality to national walfare has been the lack of accounting for environmental amenities in a system of national accounts (gross domestic product, for example). This topic is reviewed by Beattie in chapter 2.

2

Figure 1

Trade and environment nexus

I--Policy purpose II--Policy instrument III--Policy effect

r.ghi#'~kiigi!WEi.WiJ!iI~;&Jtri#~'t ..... ~~::~:;m.n' ...:.,",.,;.,,:m r::8W:,:g@j'MWMJJ.rri::;/,::::t"i:,?::r:: __ I'lf::fr~dli':::::::::::::::'::::::::::::::'::::::::::::::::::::::::::::~::::::::::::::.:::::::}:::::::I :.: :::.:.:.::::::.:' .........., . '::.... :,:: .... ::;:: ,.,.. ...... :::~~~.f}E.ri~i~~:~:m:~:~:t}~::.:}~:}~::':~:}~:::::~:}~:::::~S::::~:~}~:~:::~:

for a specified improvement in trade, or how much loss in trade is acceptable for a specified environmental protection. An example of the first type of issue is how much rainforest depletion can be justified by land clearing for soybean exports from Brazil. An example 'of the second type is how much trade loss can be justified by tighter standards on pesticide residue. The science of economics is silent in judging the relative merits of public objectives, although it can offer tools that facilitate the debate.

While it is widely accepted that freer trade leads to increased economic efficiency and economic growth, these gains neither guarantee improved environmental quality nor necessarily create environmental degradation. The effect of freer trade on agricultural resources and environmental quality depends on several critical factors that include changes in production levels, variable input use, land use, technical change, whether polluting goods are imported or exported, consumed or produced, and whether a country is large enough to affect changes in world prices or is a pricetaker. The effect of freer trade also depends on the current conditions of agricultural resources. For a country with a relatively resilient resource

base, freer trade could have little impact on environmental quality.

The second column in figure 1 represents the issues concerned with choosing between policies that rely on trade measures and those that rely on environmental controls. Policy instruments based on trade include quantitative restrictions, such as quotas and bans on imports, and' price wedges created by import tariffs or export taxes. Policy instruments based on natural resource use include regulatory approaches, such as quotas and bans on pollutant discharge, and incentive approaches like a tax on pollutant emission. ,

Environmental instruments used for environmental purposes can affect trade and environment quality. Environmental policies and regulations affect cost structure by altering inputs or levels of productivity. The argument that increased costs associated with environmental protection leads to a loss of competitiveness is most often stated in the context that countries with less stringent environmental regulations than others may gain a cost advantage, or that tighter unilateral environmental management on the part of one country places it at a cost disadvantage. In the

3

long run the fear is that pollution-intensive industries would migrate to less regulated settings.

Cells (A and C) and (8 and F) represent relatively straightforward situations in which the policy purpose (trade and environment, respectively) is treated with policy instruments that function in the same sector. Cells (A and D) and (8 and E), however, represent situations in which one goal or policy purpose is affected by policies or policy instruments designed largely to meet other goals (trade purpose - environmental instrument; environmental purpose - trade instrument). This raises the likelihood of unintended consequences and complicates the process of policy design. Nonetheless, such policy designs that cut across sectors are common. For example, environmental purposes are explicitly served in the General Agreement on Tariffs and Trade, and in the North American Free Trade Agreement. Trade instruments have been applied to nontrade purposes because they are relatively effective in the demanding task of influencing the behavior of sovereign nations.

The third column in figure 1 represents trade and environmental effects of policies. The trade effects (cells G, I, K and M) constitute the part of trade performance that is influenced by government activity. Similarly, the environmental effects (cells H, J, Land N) constitute the portion of environmental quality that is influenced by the government. This framework distinguishes the uses of "trade" and "environment" in policy analysis anci provides some context to the relationship among the articles in this report.

The issues that this report deals with roughly follow three paths through figure 1. The environmental impacts of GATT and NAFT A can be characterized as movement from A to C to H. Trade impacts of the EC Nitrate Directive or a pesticide ban move from 8 to F to M. Mixed trade and environmental impacts of bans on ivory or tropical forest products follow a path from 8 to E to K or L. Finally, the possible trade impacts from the use of environmental policies and trade measures such as nontariff barriers might also be traced from 8 to E to K.

Contents of Present Report

Countries generally pursue trade and environmental policies to augment community welfare or to increase the wel1are of a segment of the population. National income accounting is the

mO!lt often used approach in measuring welfare and gross domestic product (GDF). However, GOP does not tell us about changes in economic welfare, a concept that encompasses both economic goods and bads generated. Tobin and Nordhaus pioneered a measure of economic welfare that includes certain economic bads, such as pollution, litter, congestion, and noise, which are subtracted from GDP. A measure of well-being also may subtract depletion of our natural nonrenewable resources, such as fossil fuels and forests. In chapter 2 of this report, Beattie examines methods for measuring environmentfll value across an economy. These values are measured without regard to how policies affected them, but the measures are useful in policy evaluation.

Linking environmental value to policy design requires models of human response to policy and ecological response to human activity. Such models and the data to support them are insufficiently developed for analyzing national programs. An, approach to explicit measurement of environmental effects from policy is discussed by Mabbs-Zeno and Antle in chapter 3.

Research on policies in various countries benefits from a consistent approach to description of those policies. Policies based on trade instruments have been extensively described and compared in support of various trade negotiations. Policies based on natural resource controls are more poorly documented. In chapter 4, Gray and Vocke provide a typology of country natural resource policies, and illustrate them with country examples. International programs and agreements with environmental purposes are surveyed by MabbsZeno in chapter 5. International trade agreements are further assessed in chapter 6 by Skully, and their design for environmental purposes is examined.

The economic analysis of trade and environmental issues centers on the relationship between purposes and effects of alternative policies. Policy evaluation generally consists of quantifying the effects of a policy and comparing them to the effects of en alternative policy. The basic theory underlying analysis of the trade-environmental nexus is discussed in chapter 7 by Gray and Krissoff.

The next several chapters examine specific cases relating to particular regions. Three papers have a Western Hemisphere focus. Anderson focuses on

4

the potential environmental effects of a trade agreement among North American countries in chapter 8. Ballenger, Beattie, and Krissoff (chapter 9) appraise the effects of a wider trade agreement in the Western Hemisphere, emphasizing the tradeoff in policy design between trade and environmental goals. Forsythe and Evangelou (chapter 10) examine the trade and environmental effects from alternative means to control pests brought in with imported fruit and vegetables. The motivation for each policy is environmental in that each is intended to protect against agricultural pests. One policy, however, endangers the ozone layer of the atmosphere through its chemical emissions. The policies, therefore, offer a tradeoff between ozone protection and commodity trade.

Empirical studies in this report also examine cases in which the effect of environmental policies on competitiveness is a major concern in the European Community. Leuck lchapter 11) relates a policy with environmental purposes, the EC nitrate initiative, to its production consequences. The following chapter by Haley relates the effect of changes in EC production to trade. Haley also considers scenarios which combine the nitrate initiative with reform of the EC Common Agricultural Policy. This pair of studies demonstrates methods for analyzing a politically important class of issues while addressing a specific case with a large impact on international trade.

In chapter 13, Giordano focuses on Malaysian forest policy. Like Haley and Leuck, he shows that domestic environmental policies can have a significant impact oli tr"de.

Darwin, Lewandrowski, McDonald, and Tsigas (chapter 14) report on the effl'!ct of all existing policies on a particular environmental attribute, global temperature, and on the effect of that attribute on agricultura! production and trade. In this case, environmental feeaback on production allows policy to affect production twice, once directly and again through its effect on the environment.

Together, the articles in this report provide a recent benchmark in economic analysis of agriculturally related trade and environment issues. They sometimes roughly quantify trade effects of policies with environmental purposes or environmental instruments, but they are less specific on environmental effects of trade policies, however defined. Methods and data for analyzing environmental effects, including feedback on production and trade, are demonstrated. Although concern over the trade and environment nexus is not new, the present direction of research offers new promise for quantifiable and reliable policy analysis. It represents a step in the direction of understanding the issues related to trade and environment.

References

Garvey, Tom. "The EC Commission on Trade and the Environment," in Agriculture and the Environment - Conflict or Cooperation edited by Caroline T. Williamson, International Policy 'Council on Agriculture and Trade, Washington, DC, 1993, pp. 212-217.

Gwartney, James, and Richard Stroup. Economics: Private and Public Choice, Second Edition, Academic Press, New York, 1980.

Lean, Geoffrey, Don Hinrichsen, and Adam Markham. Atlas of the Environment, PrenticeHall Press, New York, 1990.

Low, Patrick, and Raed Safadi. "Trade Policy and Pollution," in International Trade and the Environment edited by Patrick Low, Wurld Bank Discussion Paper No. 159, Washington, DC, 1992.

Smith, Michael. "A Trader's View on the Environment," in Agriculture and the Environment-Conflict or Cooperation edited by Caroline T. Williamson, International Policy Council on Agriculture and Trade, Washington, DC, 1993, pp. 212-217.

Tobin, James, and William Nordhaus. "Is Growth Obsolete?" in Economic Growth; National Bureau of Economic Research General series, No. 96E, Columbia University Press, New York, 1972.

Tyers, Rod, and Kym Anderson. Disarray in World Food Markets: A Quantitative Assessment, Cambridge University Press, Cambridge, Engiand, 1992.

5

Chapter 2

Environmental Accounting: Including the Environment in Measures of Well-Being

Rachel Beattie

Unlike most economic assessments of welfare, this paper compares economic measures of living standards across countries, including an evaluation of environmental degradation. Efforts to include price measurements of "environmental depreciation" in the System of National Accounts are detailed, in response to charges that traditional measures of economic growth undervalue the natural resource base and overestimate social welfare.

Introduction National Income Accounting

The national income accounts originated in the The examination of trade, both current trade 1930's, influenced by the Keynesian concerns withpatterns and future trade agreements, raises the the business cycle and unemployment. Thequestion of the effect of trade on economic wellSystem of National Accounts, which began in being. Most trade research has used conventional 1942, has several measures of national production: measures of economic welfare, such as producer gross national product (GNP), gross domestic and consumer surplus. Trade also affects income, product (GDP), net national product (NNP), and usually measured by the System of National disposable personal income (DPI). In the 1946Accounts (SNA). These accounts typically cover publication, Value and Capital, Hicks emphasized only goods and services that are exchanged in the importance of net income, writing that it ismarkets while omitting nonmarket-based activities; more relevant than other measures of economicthus, the accounts may neglect important aspects well-being because it represents the amount thatof welfare. For exampltl, these accounts have long society consumes after maintaining society's stockbeen criticized for excluding nonmarketed labor, of capital:such as work within the home or subsistence

agriculture. More recently, some have argued that The purpose of income calculations in practical these longstanding measures are faulty because affairs is to give people an indication of thethey omit the effect of environmental externalities amount which they can consume withouton current well-being and the effect of natural impoverishing themselves. Following this resource depletion on future well-being. idea, it would seem that we ought to define a man's income as the maximum value which heThis paper examines efforts to modify the System can consume during a week and still expect toof National Accounts with price measurements or be as well off at the end of the week as henonprice indexes of environmental degradation-was at the beginning. Thus, when a person efforts referred to as "environmental accounting." saves he plans to be better off in the future;Some environmental accounting approaches focus when he lives beyond his income he plans toon measuring changes in the natural resource base, be worse off.such as depletion of fossil fuels or clearing of

forests. Other appr:oaches attempt to measure However, net income in conventional national environmental externalities. Examples of these accounts omits the depreciation of several sources externalities include various types of pollution or ':If capital. Because of severely underutilized contamination that lower the standard of living: capacity in the 1930's, the Keynesians neglected litter, noise, toxic waste, air pollution, and so the contributions of capital embodied in natural forth. They also include environmental amenities resources to a nation's economic welfare. The NNP' ,that enhance welfare, such as recreation, national corrects partially by including the deterioration of .parks, and scenery. Environmental accounting such capital goods as plants and machinery by'may assist our analysis of trade or economic subtracting a yearly depreciation allowance, bl!tgrowth by drawing a more complete picture of does not allow for the degradation of naturaleconomic well-being. resources.

6

Accounting for Resource Depraciation and Depletion

Environmental economists have attempted to correct the asymmetric treatment of produced capital and natural resources by including accounting for natural resource deterioration andlor depletion. The main efforts at environmental accounting have focused on two environmental effects: changes in environmental quality cl3used by production (for example, pollution effects) and change in the value of stocks of natural resources (depreciation/depletion). Research in th0 United States and Japan has focused on the former, while research in resource-based economies (the Scandinavian countries and many less developed countries) has concentrated on the latter.

Two main approaches to environmental accounting have been developed. One group advocates developing an account auxiliary to the existing SNA accounts to monitor physical changes in the environment. These accounts would have a measurement or index of environmental degradation in nonmonetary terms. Another school emphasizes the importance of measuring a price for the environment because that price may simplify comparing environmental changes with other statistics that use money changes. Some would have that monetary account integrated into the SNA to create a measure of "sustainable" national income, while others would maintain a monetary environmental account separate from GNP.

Physic81Accounung

The n"nmonetary or physical accounting branch of environmental accounting has concentrated on creating separate accounts that map the flow of such physical resources as material goods and energy. The opening balance or stock of goods is recorded as input, and the closing balance is recorded as output. The balance could change due to several factors: degradation or use, natural growth, andlor the discovery of resources or new reserves. In theory these accounts would be thorough in coverage and would show the transformation of natural resources into goods and into pollution. This approach has been attempted in Norway and also in France, which has the most inclusive and ambitious approach: Les Comptes du Patrimoine Naturel ("patrimonial accounting"). In France, the term has a broader application, also including cultural sites. These physical accounts list the sources of environmental goods on one

side and the uses of environmental goods on the other side (Peskin, 1991).

Monetary Accounting

Advocates of monetary environmental accounts attempt to estimate prices for environmental goods. In low-income countries these efforts have focused on valuing natural resource depreciation/depletion. The effect of resource depletion in these economies can be significant because many of these countries rely heavily on natural resources for export revenues, employment, and production.

There is concern about the methodology for measuring depletion and the loss in future income due to the depletion. For example, consider the case of an oil-exporting country that is depleting its reserves. Currently under the SNA, 100 percent of the oil sold counts as net income, despite the fact that the reserves are not permanent. Yet, net income &s defined by Hicks implies sustainability, and income from the reserves is not sustainable. Some propose the opposite extreme of not counting oil seld as income; instead, they advocate that all profits from depletable resources (such as oil) represent depreciation. However, under that system, a country that never used its mineral deposits (zero depreciation) would have a higher sustainable or net income than a country that did use its deposits (positive depreciation). The second country does get some benefit (income) from the deposits (or reserves) that the first country does not get because the first does not use the deposits. For the first country, the deposits would have no value in the present and in the future. The second country would more accurately have a higher income. (For example, Saudia Arabia surely has a higher net income if it uses its oil deposits than if it does not.) Also, a country that uses a depletable resource may invest the income in a renewable income source and have sustainable income in the future. Hartwick and Hageman (1992) argue that one could sell a mine, for example, and deposit the sale money in the bank to receive an infinite stream of future income.

Under this interpretation, depletable resources represent both income and depreciation. The debate centers on how to measure what percentage represents income and what percentage depreciation. Hartwick and H8geman recommend calCUlating depreciation by measuring the change in the value of reserves (a mine, in

7

their example) over time. However, if stock market data on extractive firms are not available, the researchers must estimate the value of the firm, a complicated and potentially controversial exercise.

Hartwick and Hageman describe an alternative approach called the "total Hotelling rent" method. The method estimates the quantity extracted and the profit on the marginal ton to calculate the depreciation. Alternatively, the method advanced by EI Serafy (1991) entails estimating the quantity extracted during the current period, the current total profit and the remaining reserves to estimate depreciation. '

The Role of Technology and Discoveries of New Reserves

All the methods described heretofore have the weakness that they rely on the current worth of the depletable resource. Future worth is uncertain, however, because technology can change. For example, if in the future, a new technology were developed that made generating renewable energy cheaper than extracting fossil fuels, then oil reserves would have diminished worth as an energy source. Also, new reserves may be discovered that may change the worth of the deposits.

There is the issue of when new discoveries of reserves should be included in income. (In ~r (itrast, additions to produced capital, or investment, occur and are recorded in a definite time period.) It could be argued that, if depletion of reserves is considered and subtracted from net income, discovery of new reserves should be included as income in the year of its discovery, with possibly a discounted value of future income flows.

Case Studies Incorporating Environmental Accounting

Robert Repetto and his colleagues at the World Resources Institute (WRI) have focused on depreciation of many types of natural resources (Repetto and coauthors, 1989). In a study on Indonesia, the Repetto team estimated that the annual income growth between 1971 and 1984 was 3 percent lower than if the effects of natural resource depreciation were included. They found that soil losses amollnted to 40 percent of the net value of crop production and that net losses of

'For a more detailed examination of these methods see Hartwick and Hageman (1992), pages 330-33.

forest resources were actually greater than the value of the timber harvested.

Similarly, Repetto and WRI estimated resource accounts for Costa Rica between 1970 and 1989, examining three "renewable" resource sectors: forests, fisheries, and agriculture. Combined with mining, these resource activities accounted for 17 percent of the national income, 28 percent of employment, and 55 percent of export earnings. They calculated the present value of potential future rents by discounting the value of the stream of future income and subtracting costs. The costs of soil nutrients, decline of fishing species; and loss of timber were measured, using marketed costs.

Their calculations show that the Costa Rican economy suffered a loss of 1 year's worth of GOP during the 20-year period due to natural resource depreciation. For example, in the case of forests, they subtracted the loss of forests (cleared for agriculture and for timber) from the gain in secondary forests to get net change ~n timber. In 1984, they t::alculated that the value d forests loss was $167 million, amounting to $69 for each man, woman, and child in Costa Rica. Much of the forest cleared is for agricultural use or for pasture, although often the land is not well suited to grazing.

Likewise, fish reserves experienced a net negative depreciation in the 1980's. Fisheries in the Gulf of Nicoya typify a "tragedy of the commons" as they have been depleted far beyond the economic optimum. For example, between 1985 and 1988, the quantity of white 'shrimp declined 70 percent and fidel shrimp declined 90 percent; and between 1975 and 1987 sardine production declined by 91 percent. It is difficult to measure the maximum sustainable yield because yields fluctuate naturally with weather, currents, and so forth. With limited fishing, the fish reproduction rate increases, but with more intensivs fishing, the fish may have a lower population size and may even decline until no fish are left. The Repetto team estimated the fisheries' sustainable yield, based on a measure of "fishing effort" (or intensity), and recorded yield, findinv- maximum sustainable rents occurred in the ear~ ~l'980's. The total loss between 1981 and 1989 in potential income was estimated at about 2,240 million colones. Although worth less than 1 year's loss in soil erosion, the loss is large compared with the size of the fishing sector.

8

To estimate erosion and soil loss, the Costa Rican study uses a Universal Soil Loss Equation. Assuming that soil loss can be replaced with agrichemicals, the study approximates the onsite cost of soil loss with the cost of replacement fertilizers. The offsite costs included loss of reservoir capacity (difference between real value with sedimentation and value with designed useful life) and increases in flooding capacity. They found a fairly steady depreciation of about 2,600 million colones per year between 1970 and 1989, amounting to about 9 percent of value added in agriculture.

The Costa Rican example illustrates how Government policies may encourage economic "growth" while overlooking resource depletion. Government policies have encouraged the expansion of agriculture through tax incentives, subsidies to the beef industry, and land tenure laws that emphasize "improving" land for legal title. Likewise, the Costa Rican Government has encouraged fishing through offering such incentives as tax forgiveness on inputs and preferential fuel prices. Environmental accounting might help alert policymakers to environmental problems.

Valuation of Nonmarketed Environmental Goods

Most of these developing-country environmental accounting studies examine only the marketable costs of depreciation and mineral deposit extraction, and not other nonmarketed environmental goods (or externalities) such as esthetic value, biodiversity, or pollution. In industrial nations, however, natural resources constitute a much smaller percentage of annual production than in developing countries. Thus, efforts at environmental accounting in these countries have tended to focus on measuring environmental byproducts, such as water and air pollution.

Measuring the worth of many environmental goods is complicated because either they are not exchanged in the market (such as endangered species) or the market costs do not reflect the true costs (such as hospital bills for lead poisoning versus the full cost of the illness). Thus, it is difficult to estimate how much people value them. Currently, environmental and resource data tend to be inconsistent across large areas and difficult to integrate with other data sets. Methods currently in use to measure environmental goods include direct and indirect approaches.

Direct valuation methods entail using prices of activities related to the environmental problem. For example, environmental contamination may lead to health problems, sickness, and premature death. Although morally it may be difficult to assign a value to a human life, one can measure some associated costs, such as medical expenditures, loss of earnings from abs.;:nteeism, and early death. Another example is how environmental changes to a resource may lead to productivity changes that can be measured, such as a decrease in fish catch due to pollution or an increase in dairy output due to a tree windbreak (cuts down on cow illness). Another method of direct valuation is the examination of defensive or preventive expenditures (home water filters, for example). Because direct valuation methods cover only expenses that can be measured and not nonm:trketed costs or benefits such as t.he esthetic value of a clean lake, they are seen as a minimum valuation of costs, or, in other words, a lower bound.

Indirect valuation approaches try to provide a better estimate of the value of environmental goods. Some address nonmarketed costs using "surrogate" market values: they are based on the substitution for the environmental good of a good whose value can be measured, or in other words, a marketed good is used as a proxy for a nonmarketed good. For example, one could measure the cost of an artificial fish nursery to replace a wetland (Tisdell, 1993).

Contingent Valuation

Another indirect method, called contingent valuation, uses survey methods to estimate how much respondents value an environmental benefit. Survey methods involve two approaches (Pearce and Turner, 1990; '3ishop, 1990). The first one is asking people to say how much they would be willing to pay to have a particular environmental service or to prevent a particular environmental degradation. The second is asking how much people would be willing to accept as payment for' an environmental loss or for forgoing an environmental benefit. Although, theoretically, the willingness to payor to accept would be the same, they differ in empirical tests. It has been found that people view differently losing an environmental amenity which they already have and gaining an environmental amenity which they do not yet have (or have not had recently.)

9

The main drawbacks to contingent valuation are potential biases. One bias is a "starting point" bias, which resuits from how a respondent is asked to choose among a range of options. Asking whether someone would pay between 0 and $200 or asking 0 and $100 may lead to different valuations, even if no one bids over $100. Another example is a "hypothetical bias," which arises because respondents never actually have to pay the money, and thus, may inaccurately estimate how much they would choose to pay.

Peskin's Synthesis

Another approach, forwarded by Henry Peskin, which he calls the "neoclassical" framework, attempts to include both direct and indirect costs (Peskin, 1991). He uses direct valuation techniques such as those cited above, examining costs to protect a resource. Then, he adds the mostly nonmarketed costs of any "denial of access to an environmental asset" due to pollution or degradation. In other words, other users of the asset suffer a burden if it deteriorates in quality. For example, people bear a cost from recreation losses or fishing losses in a polluted lake. The main weakness to this approach is that it relies on imprecise monetary measurements of lost benefits.

In addition to the three conventional sectors under the System of National Accounts--government, industry, and households--Peskin's accounting framework adds the environment or nature as a fourth. The adverse effects from resource degradation are treated as negative output within this category (even though industries, households, and governments may be the injured parties) and environmental services are treated as positive output. Peskin argues that this environmental account could be added to the conventional Net National Product.

A preliminary Environmental Protection Agency study on the Chesapeake Bay used the Peskin framework in an attempt to account for air and water quality, as well as the value of recreational activities. The region included Washington, DC, and the counties in Maryland and Virginia that border the Bay. The study used Bureau of Economic Analysis datai'rom the U.S. Department of Commerce on gross State product to estimate the gross county products that form the basis of obtaining a gross "Chesapeake region" product. Physical measures of water quality (nitrate end phosphorous deliveries into the Bay) and of air quality (based on measurements from the National Acid Precipitation Assessment Pr01ram) were

used. Survey methods (contingent valuation) were used to estimate the value of recreational activities such as hiking, hunting, and water sports.

The study found that in 1985 the total gross product (similar to GNP) for the "Chesapeake region" was $142.5 billion, and the net product (similar to NNP) was $125.3 billion. The environmental damages to the air were estimated to be $110 million, and to the water were estimated to be $347 million. Net environmental depreciation was calculated to !-~ $78 million. The estimate of final consumption,.. lonmarketed environmental services was a p~ .tive $1.1 billion. Thus, the net change from the environmental sector was environmental benefits minus environmental damages and depreciation. The value was a positive $568 million, making the modified net product $125.9 billion. 2

In contrast to the Peskin approach, Den Butter (1992) argues against including environmental measures into the GNP, because the net income fjgure might obscure how sustainable any growth is: "A 3-percent growth of the green GNP can, for instance, either be the result of a 5-percent growth of actual GNP and 2 percent additional pollution or a 1-percent growth of GNP and a reduction of environmental damage by 2 percent." He advocates creating a separate, nonmonetized index of environmental quality that could be used in making policy, similar to measures of unemployment, price stability, balance of payments, and economic growth.

Conclusion

In all of the environmental accounting approaches, most of the work has been theoretical; few full environmental accounts for any nation have been developed. If a standard environmental accounting approach were developed and applied, a primary benefit would be the creation of a more accurate picture of wealth and income. This set of measures could be used with other economic indicators and statistics to make comparisons across countries and across time. A drawback is that environmental accounting does not explain how the environment and economic growth cause or affect each other, nor does it describe how to counter an environmental problem. The availability

21n this example, the benefits from the environment exceeded the environmental losses, although many environmental problems were not included, such as solid and hazardous wastes and loss of biodiversity.

10

and use of these measures, however, could aid frameworks that provide for more environmental policymakers in identifying environmental problems amenities. and in designing policies or institutional

References

Adger, W.N., and M.C. Whitby. "Land Use Externalities in National accounting," in National Income and Nature: Externalities, Growth and Steady State. Edited by J.J. Krabbe and W.J.M. Heijman. Kluwer Academic Publishers. London. 1992.

Ahmad, Yusuf, Salah EI Serafy, and Ernst Lutz, editors. Environmental Accounting for Sustainable Development. World Bank, Washington, DC, 1989.

Bishop, Richard C. "The Contingent Valuation Method" in Economic Valuation of Natural Resources. Edited by Rebecca Johnson and Gary Johnson. Westview Press. Boulder, San Francisco, and Oxford. 1990.

Costanza, Robert (ed.,. Ecological Economics: The Science and Management of Sustainability. Columbia University Press, New York. 1991.

Daly, Herman, and John B. Cobb. For the Common Good: Redirecting the Economy Toward Community. The Environment and A Sustainable Future. Beacon Press, Boston. 1989.

Den Butter, Frank. "The Mirror of Cleanliness: On the Construction and Use of An Environmental Tax," in National Income and Nature: Externalities. Growth and Steady State. Edited by J.J. Krabbe and W.J.M. Heijman. Kluwer Academic Publishers. London. 1992.

Eisner, Robert. "Extended Accounts for National Income and Product," in Journal Of Economic Literature. Vol. XXVI. December 1988, pp. 1611-1684.

EI Serafy, Salah. "The Environment as Capital" in Ecological Economics: The Science and Management of Sustainability. Edited by Robert Costanza. Columbia University Press, New York. 1991.

Faber, Malte, and John L.R. Proops. "National Accounting, Time and The Environment: A NeoAustrian Approach," in Ecological Economics: The Science and Management of Sustainability. Edited by Robert Costanza. Columbia University Press, New York. 1991.

Hannon, Bruce. "Accounting in Ecological Systems," in Ecological Economics: The Science and Management of Sustainability. Edited by Robert Costanza. Columbia University Press, New York. 1991.

Hartwick, John. "Natural Resources, National Accounting and Economic Depreciation," in Journal of Public Economics, 43:291-304. 1990.

Hartwick, John, and Anja Hageman. "Economic Depreciation of Mineral Stocks and the Contribution of EI Serafy" in Toward Improved Accounting For The Environment. Edited by Ernst Lutz. UNSO-World Bank Symposium. Washington, DC, March 27,1992, p.329.

Hicks, John. Value and Capital. Oxford, Oxford University Press. 1946.

Hueting, Roefie. "Correcting National Income for Environmental Losses: A Practical Solution for a Theoretica! Dilemma," in Ecological Economics: The Science and Management of Sustainability. Edited by Robert Costanza. Columbia University Press, New York. 1991.

Johnson, Per-Olov. The Economic Theory and Measurement of Environmental Benefits. Cambridge University Press, Cambridge, Eng., New York. 1987.

Lutz, Ernst, and Michael Young. "Integration of Environmental Concerns into Agricultural Policies of Industrial and Developing Countries," in World Development. Vol. 20, No.2 pp.241-253, 1992.

11

Maler, Karl-Goran. "National Accounts and Environmental Resources," in Environmental and Resource Economics, 1:1-15, 199'1.

Pearce, D.W., and R.K. Turner. Economics of Natural Resources and the Environment. Johns Hopkins University Press, Baltimore. 1990.

Peskin, Henry. "Alternative Environmental and Resource Accounting Approaches," in Ecological Economics: The Science and Management of Sustainability. Edited by Robert Costanza. Columbia University Press, New York. 1991.

Peskin, Henry, and Janice Peskin. "The Valuation of Nonmarket Activities in Income Accounting," Review of Income and Wealth, Vol. 24, No.1, March 1978.

Peterson, George L., Beverly L. Driver, and Perry J. Brown. "The Benefits and Costs of Recreation: Dollars and Sense" in Economic Valuation of Natural Resources. Edited by Rebecca Johnson and Gary Johnson. Westview Press. Boulder, San Francisco, and Oxford. 1990.

Repetto, Robert. "Accounting For Environmental Assets," in Scientific American, June 1992. pp.71-91.

Repetto, Robert. "Earth in the Balance Sheet: IncorpClrating Natural Resources in National Income Accounts," in Environment, Vol. 34, No.7. September 1992. pp. 94-100.

Repetto, Robert, Wiiliam Magrath, Michael Wells, Christine Beer, and Fabrizio Rossini. Wasting Assets: Natural Resources in the National Income Accounts. World Resources Institute. Washington, DC, 1989.

Tisdell, Clem. Environmental Economics: Policies for Environmental Management and Sustainable Development. Edward Elgar Publishing, Ltd. Brookfield, VT. 1993.

12

Chapter 3

Measuring the Impact of Government PO~DlCies on the Environment

Carl Mabbs-Zeno and John Antle

Analyses of the impact of government support policies on agriculture have benefited from the development of quantitative measures like the Producer Subsidy Equivalent (PSE), but this measure doesn't account for costs and benefits of policies associated with the environment. In this paper, a theoretical basis is laid out for quantifying the environmental impacts ofpolicies designed principally for agricultural purposes. While much work would need to be done to put this theory into practice, it serves as a useful starting point in devising a practical measure for policy analysis. .

Introductcon

A surge of international interest in environmental problems has raised the demand for research on agricultural and environmental policies and their economic and other effects. For example, a Federal judge ruled in June 1993 that the President must have an environmental impact statement prepared on the draft North American Free Trade Agreement before it could be ratified. The ruling, however, was subsequently overturned. Such research, however, is hampered by the complexity and poor documentation of the relationship between policy and environment (Bishop and Ervin 1992). This paper presents a conceptual framework and identifies some of the research needs for quantifying the effects of policies on the environment that are designed principally for agriculture.

Problems in Environmental Policy Analysis

Analysis of environmental policies is hampered by the difficulty of quantitatively modeling or assessing policy reforms. Quantitative evaluation of environmental policies suffers from:

(1) inadequacy of data describing the environment, and

(2) poor correlation between available measures of policy level and environmental quality.

The global set of databases on the environment is being expanded by several international agencies, using a variety of strategies. The environmental monitoring programs of international organizations were recently surveyed by the U.N. Environment Programme (Fritz 1990). Most data efforts compile existing data (World Resources Institute

(WRIl 1992; World Bank 1992, pp. 196-205; Organization for Economic Cooperation and Development (OECD) 1991; U.N. Environment Programme (UNEP) 1990b). Others actively measure the environment themselves (UNEP 1990a, Fritz 1990). Still others have programs to encourage countries to expand efforts at measuring their internal environmental changes, such as the country reports submitted to the U.N. Conference on Environment and Development and the World Bank's encouragement of National Environmental Action Plans (Falloux and coauthors 1991). The national data collections are separable into natural resource accounts and state of the environment reports (Friend and Rapport 1991).

The global set of databases on policies with environmental purposes consists of a less systematic collection of items compiled in the course of developing the environmental data. Policies affecting agriculture are described for most countries by the Economic Research Service (ERS) (1988). The effects of these policies on production and consumption are also measured for many countries by ERS (Webb, and coauthors 1990) and the Organization for Economic Cooperation and Development (OECD 1992).1 Descriptions of environmental policies globally are being compiled by ERS (1992) and by the U.N. Conference on Trade and Development (UNCTAD 1992). The first stages in an inventory of policies affecting the environment have been undertaken by the Agency for International Development (USAID 1992).

Analytical tools linking policy to environmental effects are less practical than those linking policy to agriculture, trade, or many other effects that

'Both the qualitative and the quantitative inventories of policies affecting agriculture by ERS are being updated.

13

Figure 1 Relating agricultural and environmental production levels

cornacres production

acres

productive capacity .., ... N... ...

are simpler to measure than the environment. The geographic information ,systems offer new technology for integrating the variety of data used in environmental analysis, but these systems have not yet contributed substantial insights for policy design (Fletcher and Phipps 1991).

Conceptual Structure of Model

While data quality is a concern in all research, it is particularly problematic in analysis of environmental policy effects because existing models generally require data that is unavailable in most countries.2 The purpose of the conceptual model presented here is to define measurable relationships between policies and environmental quality. It is structured to compare indicators of agricultural output and environmental quality under observed or simulated policy regimes to a hypothetical regime without policy intervention.

The connection between policy elements and model outputs requires specification of production functions for each agricultural and environmental product. Figure 1 represents a case having one agricultural product (corn), one nonvalued,

2Environmental policies, as used here, refer to all policies affecting the environment.

lSD,DDD tons corn

125,000 tons corn

tons fertilizer

tons fertilizer

Angle 'pr' determined by price ratio of Inputs

environmental product (productive capacity), and two inputs (fertilizer and land).3 The upper-right quadrant and lower-left quadrant displo.'t ~he production functions for both inputs with respect to corn and productive capacity, respectively.

The production functions appear as a set ot isoquants tracing the combinations of inputs that yield a specified level of output. The price ratio between inputs determines the expansion path along which agriculture operates and the output level is determined exogenously. The resultant levels of input use are then reflected onto the environmental product function to reveal the level of environmental quality. This exercise is done once as a baseline assuming no government policy, that is, all markets are open. Then the exercise is done under a specified policy regime and the output levels are compared to the baseline.

The model is designed to represent an environmental quality response to various inputs simultaneously. Changes in the use of an input typically imply adjustments in the use of other inputs. The effect on environmental quality follows from the new input mix. The, environmental production

3Productive capacity as an attribute of environmental quality is discussed later in this chapter.

14

function represents a particular environment with specified vulnerability to changes in inputs. The functions may not describe another environment facing similar policy alternatives.

Various policies perturb the model in various ways. The production function for corn might be altered by a constraint on technology, like deep plowing. The price ratio for inputs might be altered by usevalue taxation of land or a subsidy on fertilizer. The quantity of corn produced might be altered by an export tax. The production function for productive capacity might be altered by discovery of a higher yielding variety of corn or by encouraging cover crops on fallow land.

Model Output

The model suggests the effects of policies on agricultural output and environmental quality. This formulation eXJ."llicitly renders tradeoffs between agricultural production and environmental quality. Three types of output are produced:

(1) value of agricultural products, (2) value of certain environmental products, (3) level of other environmental products.

Agricultural output is estimated for each commodity and each nation according to the monetary value of production under the policy regime being investigated. The sum of these values can be compared with the value of production in an open market to measure the policy effect on agricultural production. This is a conventional policy-modeling exercise.

The value of environmental products cannot usually be convincingly reduced to a single number. Some, like recreational use, can be valued through well-established methodologies. Others, like biodiversity, have not been effectively linked to a monetary value. Those that can be valued are treated in the present model as agricultural output is treated and then added to agricultural output, yielding net effect of policy on all products that can be valued in monetary terms.4

The list of potential environmental products is long, but its analysis is made practical by identifying a few environmental attributes that effectively represent all remaining products. The effect of policy on these attributes is measured in terms of the attribute without reference to monetary value.

'The components of environmental quality that can be monetized are discussed by Beattie in chapter 2.

For example, biodiversity under the Endangered Species Act might be measured in terms of the number of species on earth, and compared with the number of species likely without the Act. The challenge met by the present model is to limit the list of environmental products to a useful number.

The set of nonvalued environmental products used by the model should contain few enough elements that a policymaker can simultaneously grasp the tradeoffs among all of them. It should also be comprehensive, so that all environmental p,(fects are represented. Each element should doscribe unique environmental attributes, so there is no redundancy to complicate the policymakers' evaluation.

The problem of specifying parameters for environmental effects, whose value is difficult to measure, adds to an already lengthy list of such parameters. In addition to the value of output, policy effects include risk, equity, timing of returns, and other effects on social welfare.

Policy analysts might utilize some imprecisely estimated social utility function in order to balance these effects. This is rarely effective with policymakers, who are skeptical of weighting schema for social goals. In a particular problem, however, analysis may be facilitated because some of these are found to be unimportant. For example, the available policy options may have the same impact on output and differ mainly in the risk they pose. If the list of parameters is short, and easily understood and transparently linked to concerns of policymakers, a formal procedure for aggregation may be unnecessary. The analyst offers enough information for policy decisions by showing the tradeoffs associated with policy alternatives.

It is proposed here that five areas should be addressed: biodiversity, global temperature, human health, productive capacity, and maintenance. Table 1 lists several environmental problems related to each of these five categories. The selection of these categories carries the assumption that they will, in some way, be compared with each other and to agricultural output outside the model. The categories were also selected to be measurable with available data.

The table also suggests units for measuring each environmental attribute. These units are suggestions that can be accepted only after effective use in empirical study. They generally

15

Table 1--Policy effects on the environment

Attribllte Potential effects of agricultural policy Measurement unit

Biodiversity Habitat destruction for agriculture, pollution of wildlife habitat, risk from promulgation of few genotypes

Number of endangered species

Global temperature

Methane production from cattle and rice farming, forest clearing for agriculture

Global CO2 level

Human health Pesticide residues on food, worker safety, Life expectancy change weighted by water quality number of people affected

Productive Erosion, fertility loss, desertification, capacity nonagricultural production processes

(forestry and fisheries)

Maintenance Irreversible parts of productive capacity loss

are used to compare a part of the environmental situation under a specified policy, whether present, past, or prospective, with the open market situation.5

Biodiversity

Biodiversity can be measured by the number of species in the world. Therefore, the policy effect can be measured as the number of species with the policies compared with the number under an open market. This measure' could be strengthened by accounting for risk, perhaps by adding the difference in number of endangered species in the two policy regimes. It fails to account for the risks from monocropping, although they might be captured by open market studies if sufficient information were available.

Global Temperature

The contribution of a policy to global temperature could be represented by global CO2 level. This is the main "greenhouse" gas associated with agriculture. A policy with known effects on deforestation, rice production, and cattle production could readily be expressed in terms of effect on global CO2 , Policies whose main effect is on methane production, for example, could be expressed on CO2 terms, based on the relative contribution of methane and CO2 to global warming.

5The open market is used as a point of comparison because it is well understood, and is often approximated by observable experience.

Value producible by best use best use immediately after fully instituting policy, without cost of adjusting to best use

Average annual change in productive capacity following adoption of policy

Human Health

The effects of a policy on life expectancy within a country can be measured. This is biased, however, toward policies affecting younger people and mortality. It may also be possible to obtain measures of mortality or morbidity more directly associated with agricultural activities and policies, such as cases of pesticide poisoning or cases of water-borne disease.

Producfive Capacfty

Productive capacity is the central interest in evaluating agricultural policy effects. For the purpose of environmental impact, it must be distinguished from the effect of policy on production quantity. It should not, then, be interpreted as how much would be produced with the policy in place ,compared with how much would be produced in an open market. That is not an essential environmental concern. Rather, this measure should report the ability of natural resources to contribute to production under a specified policy compared with an open market situation.

The ac:tual level of production need not correspond to the capacity. It might differ due to other policies, inefficiency, capital constraints, and so forth. For example, with a land set-aside program, production falls and the production of the Nation is lower as long as the program is retained. Productive capacity, however, in the sense of what could be produced under the most productive

16

Figure 2 Policy effect on maintenance

value

VO 1-----===1-==--__ with open market

V11-------+--__________-r________~__~

V2

T1 T2

practices, is !If'!: lest on the set-aside land. Under another pohc.:y, capacity might fall due to intensified farming on the land in use. Again, actual production does not correspond to productive capacity.

In theory, capacity depends on frontier production functions for the commodities that could be produced and the selection of commodities to produce. The.concept can be made operational by measuring the maximum value of production possible after a policy is in place with technology and prices constant, except as affected by the policy. The price ratios among commodities determine the ratio of production while total production is constrained by resource availability.

Maintenance

Maintenance refers to the change in productive capacity that occurs after a policy has been instituted. It is the dynamic associated with productive capacity and, thus, must be expressed as a flow in contrast to the stock of productive capacity. For example, a policy that exacerbates erosion might not reduce productive capacity immediately in an area with substantial soil reserves. Eventually, however, capacity would decline.

with policy

T3 time

Maintenance and productive capacity are distinguished in figure 2. Two paths through time are shown for productive capacity. The open market is shown to result in a steady decline in capacity as some productive element of the environment degrades. For example, soil depth might be steadily declining through erosion, constraining the root zone. The policy under study is introduced at T,. It initially raises productive capacity (T2)' but eventually reduces capacity even below what it would have been under an open market (T3). In the example, this might occur with introduction of tractors on a tropical soil. At first, the machinery deepens the root zone, offsetting erosion, but eventually it damages the soil through compaction. Productive capacity was raised relative to the open market, observed shortly after the policy took affect. Maintenance was lowered since the changes in expected value of productive capacity were more negative with the policy than with the open market.

This set of nonmonetized environmental attributes encompasses most of the major effects on the environment that currently concern policymakers. Health issues cover the main externalities that cannot be convincingly monetized.6 Global

6Externalities are effects that producers or consumers impose on others without opportunity for compensation.

17

warming and declining biodiversity are the main transboundary effects of national policies at a global level. The other two attributes represent the quality of the environment in producing commercial services. Productive capacity represents the stock of capacity while maintenance represents changes in the flow of services.

Comprehensive coverage of relevant policy effects remains a problem in this framework mainly because the measures of environmental attributes are weak rather than because there are too many attributes. As more data are collected and as better methods for relating various aspects of an attribute to the same measure are developed, the coverage of the model will improve. For example, human health and longevity are affected by the environment. The correlation between these might be better understood with better data, leading to a better measure of health than the simple life expectancy measure recommended here.

Redundancy among the attributes listed remains a problem because components of nonmonetized environmental attributes are measurable in monetary terms. For example, in studying a particular country, the monetized products might include agricultural output and recreational value. if an endangered species living in area contributes to recreational value, the policy effect on recreation must be carefully distinguished from the effect on non monetized biodiversity.

The short length of this list is a tribute to the effectiveness of the science of economics in placing values on externalities, although the presence of the list at all is a recognition of the limitations of valuation exercises. It should be possible to effectively represent the circum'&tances of each country without losing comparability with studies of other countries.

Research Needs

To apply the model, the relationships between input and products must be specified. Agricultural production functions are often estimated for policy analysis. Environmental production functions are occasionally used, but the measure of environmental qUl3lity varies widely. The major gaps in existing knowledge for use of this model are the production functions for specific, nonmonetized environmerltal services.

A set of environmental production functions could be determined without reference to a specific policy question, but the task of assessing the vulnerability of each country in the world to five components of environmental quality would be daunting. Rather, the experience gained from ongoing policy analysis might be eventually collected, and a classification of countries into categories of similar vulnerability might be used to build a global model. If policy research adopts a consistent set of policy measures for environmental effects, as suggested here, the task of comparing experience in various countries would be enhanced.

References

Agency for International Development, U.S. Department of State. "A Policy Taxonomy and Analysi.s of Policies Affecting Natural Resources and the Environment," APAP II, No. 406, 1992.

Bishop, Richard, and David Ervin. "New Directions in Data Information Systems and their Uses: Natural Resources and the Environment," Proceedings of the Symposium on New Directions in Data and Information for Rural Areas, Ames: Iowa State University Press, 1992.

Econ. Res. Serv., U.S. Dept. of Agri. Global Review of Agricultural Policies, Staff Report No. AGES880304, 1988.

Econ. Res. Serv., U.S. Dept. of Agri. Global Review of Environmental Policies: Research Plan, mimeo, 1992.

Falloux, Francois, Lee Talbot, and Jeri Larson. Progress and Next Steps for National Environmental Action Plans in Africa, Washington, DC, June 1991.

Fletcher, Jerald J., and Tim T. Phipps. "Data Needs to Assess Environmental Quality Issues Related to Agriculture and Rural Areas," American Journal of Agricultural Economics, August 1991, pp. 927-44.

18

----

Friend, Anthony M., and David Rapport. "Evolution of macro-information systems for sustainable development," Ecological Economics, 3(1991), pp. 59-76.

Fritz, Jan-Stefan. "A Survey of Environmental Monitoring and Information Management Programmes of International Organizations," Munich: U.N. Environment Programme, 1990.

Organization for Economic Cooperation and Development. Environmental Indicators, Paris, 1992.

U.N. Conference on Trade and Development. "Reconciliation of Environmental and Trade Policies," unpublished manuscript, 29 June 1992.

U.N. Environment Programme. GEMS: Global Environment Monitoring System, Nairobi, 1990a.

. GRID: Global Resource Information Database, Nairobi, 1990b.

Webb, Alan, Michael Lopez, and Renata Penn. Estimates of Producer and Consumer Equivalents: Government Intervention in Agriculture, 1982-87, Statistical Bulletin No. 803, Econ. Res. Serv., U.S. Dept. Agr., 1990.

World Bank. World Development Report: Development and the Environrt:lent, 1992.

World Resources Institute. Environmental Almanac, Boston: Houghton-Mifflin, 1992.

. World Resources: A Guide to the------Global Environment, New York: Oxford University Press, 1991.

19

Chapter 4

Policy Choices in the Use of Land and Water

Denice Gray and Gary Vocke

The objective of this paper is to describe through country examples (1) allocative systems that determine who can use the land and water resources, (2) public control or influence over allowed uses of these resources, and (3) public protection of these resources. This paper describes the allocative systems and public policies affecting resource use of various countries. These systems and policies differ for many reasons, including history of the country and scarcity of the resource. The systems and policies that have evolved for land and water are different because land stays in one location while water flows from one location to another, and because of the uncertainty of the water supply.

Introduction1

As mentioned in the introduction to this report, natural resource policies to manage land and water are environmental instruments with both environmental and trade effects. How land and water resources are managed influences agricultural production around the world and, hence, trade among countries. For example, policies to expand irrigation can increase yields and production. In Indonesia, the development and upgrading of public irrigation projects for growing high-yielding rice varieties quickly moved the country from being the world's largest rice importer to self-sufficiency. Policies to managfl agricultural practices to protect land and water quality can also affect production and trade. For example, the U.S. Conservation Reserve Program, which had environmental improvement as a primary objective, took highly erodible and environmentally sensitive cropland out of production, reducing the avai!ability of grain for export.

Land and water are crucial inputs into agricultural production. These resources also have many competing uses in other sectors of the economy, including scenic beauty and wildlife habitat. Societies have developed various means to allocate land and water resources among the many competing uses, and to change that allocation when needed. Allocative systems are the means by which individuals or entities become resource users. For example, many countries allow private ownership and transfer of land; thus, one can

1 Most country examples are drawn from two forthcoming ERS publications, Global Review of Resource and Environmental Policies: Land Resource Development and Management, and Global Review of Resource and Environmental Policies: Water Resource Development and Management.

become a land user by buying the land. On the other hand, in some countries the water belongs to the state, and one can only become a water user by receiving a permit from the state. The state has an interest in the specific uses made of these resources, as well as in the prevention of excessive resource degradation. Governments can achieve public goals concerning land and water use through state ownership, mandatory controls, economic incentives, or economic disincentives.

Systems of Allocative Rights for land and Water Use

Systems of allocative rights define legal uses of the resource, owners of rights, rules of transfer, and length of tenure. Property rights affect both economic and environmental incentives. That is, the level of control over the resource by users determines the extent to which users employ the resource efficiently, invest in conservation, and maintain quality.

Land Tenure and Transfer

Land tenure systems range from complete state contrel over land use, to shared control by communities, to complete control vested in individual owners. Land allocation systems are illustrated